PROVIDED. Vision is useful because it informs us about physical properties of the environment. In the case of color, one important function is to provide information about object identity - color isgenerally considered a perceptual correlate of object surface reflectance. For color to be a reliable guide to object identity, the perceived color of an object must remain stable across variations in the scenes in which it is viewed. This requirement is non-trivial because the light reflected to the eye varies with .scene factors that are extrinsic to the object, such as the.spectrum of the illumination and the reflectance of nearby objects.' The visual system adjusts to stabilize (partially) object color appearance against changes in these extrinsic factors, a phenomenon called color constancy. This proposal consists of experiments designed to study object color appearance and its constancy. Golor constancy is an example of a larger class of perceptual constancies (e.g. size constancy and shape constancy) that together allow us to perceive a stable physical world. As such, a detailed characterization of the color system may provide insights that generalize beyond color per se. The. proposed research consists of behavioral experiments with human observers that will allow a functional characterization of.how we perceive object color. Questions to be addressed include a) how does the visual system integrate information across the surface of a three-dimensional object to arrive at an overall percept of object color? b) how is this process affected by variations in the material properties of objects (i.e. variations in the object's bidirectional reflectance distribution function)? c) how stable is object color perception across variation in object shape and pose? d) how does object color perception depend on geometrical and spectral properties of the illumination? e) can we model color perception within a computational framework derived from inverse optics? and f) what role do mechanisms of adaptation tapped by objective threshold measurements play in object color constancy across changes in complex scenes? This research should help us understand how color vision works in the real world and provide a foundation for clarifying the neural processing of color. . v